Quasi-Global Momentum
Improved decentralized training with data heterogeneity
Decentralized training of deep learning models is a key element for enabling data privacy and on-device learning over networks. In realistic learning scenarios, the presence of heterogeneity across different clients' local datasets poses an optimization challenge and may severely deteriorate the generalization performance. We propose a novel momentum-based method to mitigate this decentralized training difficulty.
inactive
—
entered showcase: 2021-11-04
—
entry updated: 2024-04-09
This project has not yet been evaluated by the C4DT Factory team.
We will be happy to evaluate it upon request.
Simulation
Python
Apache-2.0